- Abs
- Acos
- Acosh
- Adam
- Add
- Add
- Add Graph
- AdditiveAttention
- AffineGrid
- All
- All
- All
- All
- All
- All
- All
- All
- All
- All
- All (Index Array)
- All (Model)
- All (Name Array)
- All (Random)
- All Inputs
- All Inputs Layers
- All Outputs
- All Outputs Layers
- AlphaDropout
- And
- ArgMax
- ArgMin
- Asin
- Asinh
- Atan
- Atanh
- Attention
- Attention
- AttnLSTM
- Average
- AveragePool
- AvgPool1D
- AvgPool2D
- AvgPool3D
- BatchNormalization
- BatchNormalization
- Bernouilli
- BiasAdd
- BiasDropout
- BiasGelu
- BiasSoftmax
- BiasSplitGelu
- Bidirectional
- BifurcationDetector
- BinaryCrossentropy
- BitmaskBiasDropout
- BitmaskDropout
- BitShift
- BitwiseAnd
- BitwiseNot
- BitwiseOr
- BitwiseXor
- BlackmanWindow
- BMP File
- Cast
- CastLike
- CategoricalCrossentropy
- CategoricalHinge
- CDist
- Ceil
- Celu
- CenterCropPad
- Clip
- Col2lm
- ComplexMul
- ComplexMulConj
- Compress
- Concat
- Concatenate
- ConcatFromSequence
- Constant
- Conv
- Conv1D
- Conv1DTranspose
- Conv2D
- Conv2DTranspose
- Conv3D
- Conv3DTranspose
- ConvInteger
- ConvLSTM1D
- ConvLSTM2D
- ConvLSTM3D
- ConvTranspose
- ConvTransposeWithDynamicPads
- Cos
- Cosh
- CosineSimilarity
- CropAndResize
- Cropping1D
- Cropping2D
- Cropping3D
- CumSum
- Custom
- DecoderAttention
- DecoderMaskedMultiHeadAttention
- DecoderMaskedSelfAttention
- DeformConv
- Dense
- DepthToSpace
- DepthwiseConv2D
- DequantizeBFP
- DequantizeLinear
- DequantizeWithOrder
- Det
- DFT
- Div
- Dropout
- Dropout
- DynamicQuantizeLinear
- DynamicQuantizeLSTM
- DynamicQuantizeMatMul
- DynamicTimeWarping
- Einsum
- ELU
- ELU
- Elu
- Embedding
- EmbedLayerNormalization
- EPContext
- Equal
- Erf
- Exp
- Expand
- ExpandDims
- Exponential
- Exponential
- EyeLike
- FastGelu
- Flatten
- Flatten
- Float
- Floor
- FusedConv
- FusedGemm
- FusedMatMul
- FusedMatMulActivation
- GatedRelativePositionBias
- Gather
- GatherElements
- GatherND
- GaussianDropout
- GaussianNoise
- GELU
- GELU
- Gelu
- Gemm
- GemmaRotaryEmbedding
- GemmFastGelu
- GemmFloat8
- GlobalAveragePool
- GlobalAvgPool1D
- GlobalAvgPool2D
- GlobalAvgPool3D
- GlobalLpPool
- GlobalMaxPool
- GlobalMaxPool1D
- GlobalMaxPool2D
- GlobalMaxPool3D
- GlorotNormal
- GlorotUniform
- GPU
- Greater
- GreaterOrEqual
- GreedySearch
- GridSample
- GroupNorm
- GroupQueryAttention
- GRU
- GRU
- GRU
- H5 To ONNX
- HammingWindow
- HannWindow
- HardMax
- HardSigmoid
- HardSigmoid
- HardSigmoid
- HardSwish
- HeNormal
- HeUniform
- Hinge
- Huber
- Identity
- Identity
- If
- ImageDecoder
- Individual by index
- Individual by index
- Individual by index
- Individual by index
- Individual by index
- Individual by index
- Individual by index
- Individual by index
- Individual by index
- Individual by index
- Individual by index
- Individual by name
- Individual by name
- Individual by name
- Individual by name
- Individual by name
- Individual by name
- Individual by name
- Individual by name
- Individual by name
- Individual by name
- Individual by name
- Input
- Input
- Input by index
- Input by name
- InstanceNormalization
- Int64
- Int8
- Inverse
- JPEG File
- JPEG2000 File
- Keras To ONNX
- KLDivergence
- L1
- L1L2
- L2
- LayerNormalization
- LayerNormalization
- LeakyReLU
- LeakyReLU
- LeakyReLU
- LecunNormal
- LecunUniform
- Less
- LessOrEqual
- Linear
- Linear
- Load ONNX File
- Log
- LogCosh
- LogSoftmax
- LongformerAttention
- Loop
- LpNormalization
- LpPool
- lrfft
- LRN
- lslnf
- lsNaN
- LSTM
- LSTM
- LSTM
- MatMul
- MatMulBnb4
- MatMulFpQ4
- MatMulInteger
- MatMulInteger16
- MatMulIntergerToFloat
- MatMulNBits
- Max
- MaxPool
- MaxPool1D
- MaxPool2D
- MaxPool3D
- MaxPoolWithMask
- MaxRoiPool
- MaxUnPool
- Mean
- MeanAbsoluteError
- MeanAbsolutePercentageError
- MeanSquaredError
- MeanSquaredLogarithmicError
- MeanVarianceNormalization
- MelWeightMatrix
- Merge
- MicrosoftDequantizeLinear
- MicrosoftGatherND
- MicrosoftGelu
- MicrosoftGridSample
- MicrosoftMultiHeadAttention
- MicrosoftPad
- MicrosoftQLinearConv
- MicrosoftQuantizeLinear
- MicrosoftRange
- MicrosoftTrilu
- MicrosoftUnique
- Min
- Mish
- Mod
- MoE
- Mul
- MulInteger
- MultiHeadAttention
- Multinomial
- Multiply
- MurmurHash3
- Name
- Name
- Neg
- NegativeLogLikelihoodLoss
- Netron Summary
- NGramRepeatBlock
- NhwcConv
- NhwcFusedConv
- NhwcMaxPool
- NonMaxSuppression
- NonZero
- Not
- One To Mult
- OneHot
- Ones
- ONNX File
- ONNX To H5
- ONNX To Keras
- ONNX To Pytorch
- ONNX To TF SavedModel
- Open Log Folder
- OptionalGetElement
- OptionalHasElement
- Or
- Orthogonal
- Output by index
- Output by name
- Output Predict
- Output Train
- PackedAttention
- PackedMultiHeadAttention
- Pad
- Permute3D
- PNG File
- Poisson
- Pow
- Predict Output
- PReLU
- PRelu
- Pytorch To ONNX
- QAttention
- QGemm
- QLinearAdd
- QLinearAveragePool
- QLinearConcat
- QLinearConv
- QLinearGlobalAveragePool
- QLinearLeakyRelu
- QLinearMatMul
- QLinearMul
- QLinearReduceMean
- QLinearSigmoid
- QLinearSoftmax
- QLinearWhere
- QMoE
- QOrderedAttention
- QOrderedGelu
- QOrderedLayerNormalization
- QOrderedLongformerAttention
- QOrderedMatMul
- QuantizeBFP
- QuantizeLinear
- QuantizeWithOrder
- QuickGelu
- RandomNormal
- RandomNormal (Generator)
- RandomNormalLike
- RandomUnifom
- RandomUniform (Generator)
- RandomUniformLike
- Range
- Raw Data
- Raw Data (Constant)
- RawConstantOfShape
- Reciprocal
- ReduceL1
- ReduceL2
- ReduceLogSum
- ReduceLogSumExp
- ReduceMax
- ReduceMean
- ReduceMin
- ReduceProd
- ReduceSum
- ReduceSumInteger
- ReduceSumSquare
- RegexFullMatch
- RelativePositionBias
- Relu
- ReLU
- ReLU
- RemovePadding
- Reset GPU Device
- Reshape
- Reshape
- Resize
- RestorePadding
- Resume
- Resume
- Resume
- Resume
- Resume
- Resume
- Resume
- Resume
- Resume
- Resume
- ReverseSequence
- Rfft
- RNN
- RNN
- RoiAlign
- RotaryEmbedding
- Round
- SampleOp
- Sampling
- Scan
- ScatterElements
- ScatterND
- Selu
- SELU
- SELU
- SeparableConv1D
- SeparableConv2D
- SequenceAt
- SequenceConstruct
- SequenceEmpty (Generator)
- SequenceErase
- SequenceInsert
- SequenceLength
- SequenceMap
- SGD
- Shape
- Shrink
- Sigmoid
- Sigmoid
- Sigmoid
- Sign
- SimpleRNN
- SimpleRNN
- Sin
- Sinh
- Size
- SkipGroupNorm
- SkipLayerNormalization
- SkipSimplifiedLayerNormalization
- Slice
- Snpe
- Softmax
- SoftMax
- SoftMax
- SoftmaxCrossEntropyLoss
- Softplus
- SoftPlus
- SoftPlus
- Softsign
- SoftSign
- SoftSign
- SpaceToDepth
- SparseAttention
- SparseToDenseMatMul
- SpatialDropout
- Split
- Split
- SplitToSequence
- Sqrt
- SquaredHinge
- Squeeze
- STFT
- StringConcat
- StringNormalizer
- StringSplit
- Sub
- Substract
- Sum
- Summary
- Swish
- Swish
- Swish
- Tan
- Tanh
- TanH
- TanH
- TF SavedModel To ONNX
- TfldfVectorizer
- ThresholdedRelu
- ThresholdedReLU
- ThresholdedReLU
- Tile
- Tokenizer
- Toolkit version
- Toolkit Version
- TopK
- TorchEmbedding
- Train Output
- Transpose
- TransposeMatMul
- Trilu
- TruncatedNormal
- UnfoldTensor
- Unique
- Unsqueeze
- UpSampling1D
- UpSampling2D
- UpSampling3D
- VarianceScaling
- Warning Parameters
- Warning Parameters
- Where
- WhisperBeamSearch
- WordConvEmbedding
- Xor
- ZeroPadding1D
- ZeroPadding2D
- ZeroPadding3D
- Zeros
Architecture
Deep Learning Toolkit / Architecture — Documentation. Browse 518 Graiphic technical references.